In French, there is a very specific phenomenon called "liaison", where words may be pronounced differently when combined with another word.

For Spanish, let's take the letter "d" as an example.

There is a strong "d" (more like the english one), and there is a soft one (more like "th").

When a words begins with "d", or after "n" and "l", the rule says it's a strong one.

In every other case, it'll be a soft one.

Now, here's the question:

This is a strong "d": dos.

This is a soft "d": cada.

Now given this sentence:

cada dos semanas

What is the second "d"? Is it a strong one, or a soft one? I'm a bit confused. I'm almost sure that when combined... what I'm hearing (at Granada, Spain) is a soft sound - although I haven't read any such rule anywhere...

Does this also happen with other consonants? E.g. "r", "b"/"v"?

  • 1
    personally i don't think there's a rule about difference pronunciation of the letter d, in Spain there are rules for "C", "Z" and "S" , but those are ignored in some parts of America, especially in Mexico, each region has different ways of pronouncing letters, examples are "CH" "LL", "Y" and the forever defying "X"
    – Mike
    Commented Jul 5, 2018 at 18:56
  • 1
    A los niños españoles en las escuelas no les enseñan variantes alofónicas de la "d", ni nada por el estilo. Les enseñan a pronunciar solo un sonido de "d", a aplicar igual a todas las vocales ("da de di do du") y con eso "van que chutan". Las variaciones surgen luego de forma inconsciente, por simple economía al vocalizar rápido, pero si les pides que pronuncien las palabras lentamente, marcando las sílabas (ca-da, da-do, etc...) solo oirás la versión "fuerte". Pero intenta usar solo el sonido "fuerte" hablando rápido. Sonará como la voz un robot (si es que lo logras).
    – Fran
    Commented Oct 5, 2018 at 23:09
  • Concuerdo con @Fran, en el proceso y libros de lecto-escritura nunca se menciona que existan 2 sonidos para la consonante "D", se enseña un único sonido para las 5 vocales es independiente de las letras que lo rodean o la posición en la palabras. Por el comentario en otra pregunta es más un asunto académico de filología. En los países de América que hablan español tampoco se enseña ni siquiera se menciona que existan 2 sonidos para la consonantes D.
    – alvalongo
    Commented Sep 19, 2019 at 16:03
  • @Mike those rules you mentioned about C, Z and S are not ignored in "some" american countries, they are ignored in all of them. So those rules only apply in Spain. Commented Sep 19, 2019 at 16:18

7 Answers 7


Disclaimer: I am not a Spanish speaker, nor have I studied Spanish to any real extent. (I've done some self-study via textbooks and online apps/websites like Duolingo.) However, I know a bit about Spanish pronunciation/phonetics because I'm interested in linguistics and I have read some articles or web pages about it.

Spanish has many cases where the pronunciation of the end of a word changes based on the sound at the start of the following word (or vice versa). But my understanding is that the processes involved are usually not particularly similar to French liaison; rather, in most cases, the Spanish pronunciation phenomena could be thought of as "allophonic" processes.

As you've said in your question, French "liaison" is a very specific phenomenon that clearly involves phonological processes, and is not just a matter of allophony. For example, Wikipedia gives the example of "premier étage", where /ʁ/ is pronounced, even though there is no general rule of French pronunciation that absolutely prohibits vowel-vowel sequences like /ee/ (as in "crée") or /ɛe/ (as in "beignet épicé"). And this contrasts with the pronunciation of a phrase like "premier vol", where there is no /ʁ/, even though the sequence /ʁv/ is allowed in French (as in "servir").

Phonologically conditioned allomorphy in Spanish

There are a few cases in Spanish of words having special phonological forms conditioned by the form of another word: for example, the definite article typically takes the form "el" when it comes directly before a feminine noun starting with a stressed /a/ sound, but "la" before other feminine nouns. But as far as I know, this kind of thing is quite rare in Spanish. Interestingly, it seems that native speakers tend not to think of this "el" as being just a form of the feminine article; it may "feel" like the masculine article is being used in this context (see the answers to the following question for more details: Why is "agua" masculine in singular form and feminine in plural? "El agua" / "Las aguas" ¿Por qué decimos "el agua" si es una palabra femenina?).

Another somewhat similar example that I found described in this document ("There is phonologically-conditioned, but not melody-conditioned allophony"), by Tobias Scheer) is the dissimilation of the conjunction "y" /i/ to "e" before words starting with the sound /i/ and of "o" to "u" before words starting with the sound /o/.


One thing that does occur in Spanish especially at word boundaries is " synalepha": two adjacent vowels across word boundaries tend to be pronounced in the same syllable. Vowels in this context apparently behave differently from word-internal vowels in hiatus, so I think we do have to mention word boundaries in the formulation of this rule.

Phrase-level allophony

Aside from this, many allophonic processes in Spanish apply at the (prosodic) phrase level rather than at the word level, including the allophony of "d" that you mentioned. See guifa's answer to How should I pronounce the Spanish consonant 'd'? So the "d" in "dos" in "cada dos semanas" would usually be pronounced according to the same rules for pronouncing word-internal /d/ in Spanish.

My understanding is that the allophony of "b/v" and "g" follows the same general pattern: approximant or fricative allophones are used between vowels and plosive allophones are used phrase-initially or after the coda nasal. (In other consonant-consonant sequences, the rules are a little bit more complicated: for example, I have read that the heterorganic sequences "lb/lv" and "lg" are pronounced with approximant allophones, but the homorganic sequence "ld" is pronounced with the plosive allophone of /d/.)

Many parts of Spanish pronunciation vary by dialect, however, so I don't know if it's possible to give a comprehensive description of the processes that may apply. For example, the coda nasal assimilates in place to a following consonant across word boundaries in many dialects, but in others I believe it is always velar [ŋ]. I also found an abstract of an article here that talks about dialect differences in the realization of "d", "b/v" and "g" after other consonants.

I don't know too much about this following topic, but the "aspiration" of word-final /s/ is I believe another area where pronunciation may vary based on the following sound.

  • 1
    Splendid answer, welcome to Spanish Language!! You may find it interesting to browse through the linguistics tag.
    – fedorqui
    Commented Jul 6, 2018 at 9:59
  • Indeed, /b/ and /g/ follow the same rules as /d/ with regards to softening. Commented Jul 6, 2018 at 10:17
  • Welcome, sumelic. Take a look around -- this site has something for everyone, and there's no fluency threshold for participation. Commented Jul 6, 2018 at 13:14

Lenition in Spanish

  1. This phenomenon is known as lenition, and is not exactly the same as liaison in French - which is a form of external sandhi (across word boundaries) which is viable depending on the syntactic environment. In contrast, lenition in Spanish is an allophonic rule dependent on syllable boundaries - it occurs within words as well as between them - and occurs independent of syntactic environment (though it is conditioned by intonational phrase)1.
  2. The second 'd' in your example will probably2 realise as a [ð], since it occurs intervocally.
  3. Yes, in addition to /d/, there are also contextual stop vs. approximant/fricative allophones for /b/, /g/, and /ʝ/ in Spanish.

Sandhi proper in Spanish

However, there are examples of external sandhi in some dialects of Spanish:

Puerto Rican

An example of an external sandhi rule in Spanish is the deletion of word-final stressed [a] in a verb form whenever it precedes a mid vowel. This is common in Puerto Rican dialect.6 Kaisse argues that word-final stressed [a] may delete only if it is contained in a verb. This deletion rule is necessarily a sandhi rule because its structural description requires syntactic information, i.e. knowledge of the grammatical category “verb”...

The following data illustrate how the deletion rule applies in verbs, but fails in nouns and adverbs.

enter image description here

6 This deletion process must not be confused with a more general variety of /a/-deletion, which targets any unstressed /a/ before any vowel, across a word boundary, and is common in many Latin American dialects.

Western Andalusian

... in Western Andalusia, where there is no vocalic contrast between singular and plural forms, the alternation of consonants has come to have morphophonemic status in the noun (and elsewhere):

/la 'bota/ "boot"
/la φota/ "boots"

/lo de'ma/ "the rest (of it)"
/lo θelma/ "the rest (of them)"

/la 'gota/ "drop"
/la 'xota/~ /la 'hota/ "drops"


1. https://www.staff.ncl.ac.uk/i.e.mackenzie/sandhir.htm
2. In careful or emphatic speech it may realise as [d] though, see:

/b d g/ en sílaba trabada se pronuncian [β̞ ð̞ ɣ̞] pueden realizarse enfáticamente como [b d g]; obturar [o̞β̞tuˈɾa̠ɾ]~[o̞btuˈɾa̠ɾ], admirar [a̠ð̞miˈɾa̠ɾ]~[a̠dmiˈɾa̠ɾ], asignar [a̠siɣ̞ˈna̠ɾ]~[a̠sigˈna̠ɾ].

  • I had no idea about the term. Thanks for letting me know! :) Commented Jul 5, 2018 at 14:42
  • 1
    i don't think this is the case, lenition is when a word CHANGES and is not about the current pronunciation of one and by the phonetic articulation of words. i think that varies with regions.
    – Mike
    Commented Jul 5, 2018 at 19:02
  • 1
    @Mike the Spanish Wiki page for lenición explicitly describes the changes /b/ → /β/ etc as examples of said phenomenon.
    – jacobo
    Commented Jul 5, 2018 at 19:09
  • too much for me... i don't get it, probably i can't hear the differences.
    – Mike
    Commented Jul 5, 2018 at 19:16
  • 2
    @Mike - It's clear for B at the beginning of an utterance versus B in the middle of an utterance. At the beginning it's more percussive, more emphatic. In the middle it's softer, less explosive. In the case of D -- well, I never realized that there is something similar going on until I read this page, but I suppose there might be a bit of that difference going on with this consonant too.... Commented Jul 6, 2018 at 2:12

When I think of the concept of liaison in French pronunciation, I think of situations like

mes amis [my friends]

Normally, the S in les is silent. But because of the next word starting with a vowel, we do pronounce the S. (It gets a Z sound there, just as in words like liaison.)

In Spanish I can't think of a single case with such a clear change, such as in "mes amis."

The facet of Spanish pronunciation that I think comes closest to this is the fact that when one is pronouncing a series of short words in a sentence or sequence, one pronounces all those syllables with a consonants come first policy. Example:

Los ases al frente, por favor.

I made this sentence up to demonstrate. It's a little silly -- it means "Aces to the front, please." (Aces being the ones in a deck of playing cards.) The way you will pronounce this in Spanish will be

Lo sa se sal fren te, por fa vor.

This is different from the French liaison in that none of the letters' pronunciation will be different than they would be if the word were pronounced in isolation. However, the reason why I wrote the pronunciation with the "consonants first" notation will become clear to you if you sing the sentence slowly.

Here's another example:

Es un elefante --> E su ne le fan te.

That is the closest I can find in Spanish to what is going on with the French liaison.

Using the "consonants first" policy will help you achieve an authentic accent (pronunciation). It will be especially helpful to get your L forward enough.

I thought of something that is closer to the French liaison:

los violines

Under normal circumstances, the S in los is a plain, garden variety S (unvoiced). But under the influence of the V, a voiced consonant, it will get voiced, and this group of words will sound like "loz violines" -- where I'm cheating a bit and using the Z to indicate the sound of a Z in English.

(If you're not familiar with voiced - unvoiced, you can figure out the difference with slow singing. When the resonance pauses during the consonant, that consonant is "unvoiced.")

  • Good point about consonants being "re-syllabified" so to speak across word boundaries. In French, this happens in addition to liaison--it is called "enchaînement".
    – sumelic
    Commented Jul 6, 2018 at 12:30
  • While the S won't change sound, letters like D or N could end up being modified (final N is like English ng in some dialects, but when resyllabified will return to a typical N) Commented Oct 4, 2018 at 1:47
  • Moreover, there are many dialects of Spanish that aspirate syllable-final /s/ (/s/ = [h]) but do not aspirate /s/ when it's word-final and followed by a vowel (los patos = [lohpátoh] but las águilas = [laságilah]). This is rather more like French liaison.
    – pablodf76
    Commented Oct 4, 2018 at 16:56

The sounds normally characterized as voiced stops in Spanish, /b/, /d/ and /g/, each have two allophones (contextual variants), as you have noted. One is pronounced as a true stop (the one you call "strong"), while the other (the "soft" one) is pronounced as a weak fricative or an approximant, and sometimes elided (dropped). This kind of weakening is called lenition. For the most part the "strong" allophones are found at the beginning of words or after another consonant, especially nasals and liquids (/r/ and /l/). This varies with dialect, but we need not concern ourselves with the specifics.

Word-initial voiced stops remain always strong, moreover, only when the word is pronounced in isolation or with special care (e.g. when the speaker wants to emphasize them). If there is a word ending with a vowel before it, the word-initial voiced stop will be lenited just as if it were inside the word and surrounded by vowels. In your example, «cada dos semanas», the initial /d/ in dos will always be "soft", because cada is a function word that always gets attached to the next word, so it's as though you had pronounced *cadadós.

This is not liaison as in French; no new sound is appearing to link the two words (enchaînement). But depending on the analysis, it can be considered a related kind of change. French liaison makes a "hidden" consonant appear at the boundary between two words when two vowels would otherwise come into contact; Spanish lenition changes the quality of a consonant in that same environment. The effect in both cases is to soften the transition between the vowels.

There are differences, though. French liaison takes into account grammatical factors that Spanish lenition doesn't. In Spanish, if there are two vowels with a voiced stop in between and no forced pause is inserted, then the voiced stop will weaken, always.

As explained elsewhere, /b/, /d/ and /g/ all become lenited in certain contexts. When I say /b/ that's the sound that you'll find written as b or v, which are pronounced the same (no matter what anyone might tell you: they mean the same sound, always). And /g/ you can of course find written as g (as in gato, gota, gula) or as gu before e and i (as in guerra, guitarra); not the same as the g in gel or gil (which sounds the same as j). All of these can weaken to the point of disappearing altogether between vowels in some dialects and in fast speech.

There's a particular case that you might run into, having to do with "S-aspiration". The sound /s/ is sometimes "aspirated", pronounced [h] (like an English h) when syllable-final. This aspiration is also a kind of lenition or weakening, and it can also turn into elision (the sound is dropped altogether). But this weakening or disappearance is normally avoided, when the sound /s/ is at the end of a word, if the next word begins with a vowel. This is in fact rather similar to French liaison.

For example, a speaker might pronounce «los patos» as [lohpátoh], aspirating the two /s/ sounds; but in «las águilas» they will pronounce [lasáγilah] ([γ] = lenited /g/). And «los patos y las águilas» would come out as [lohpátosi lasáγilah] (since y = /i/, a vowel).

This is a very simplified explanation of actual S-aspiration in Spanish. Some dialects do not have it and others do different things with it, but that's the idea.

  • Excellent answer. I had imagined the S was the same as the N, but I hadn't paid close enough attention to speakers from aspirating areas enough to notice if indeed the aspiration was there between vowels. Interesting that y can prevent it since it more strongly links forwards than backwards. Commented Oct 6, 2018 at 15:07
  • 1
    Well, as I said, there's more to it. I've heard people aspirate word-final /s/ regardless of whether a vowel follows.
    – pablodf76
    Commented Oct 6, 2018 at 22:32

You were sort of asking several things at once. In this answer I will focus specifically on what happens with B and V. You asked what happens with the Ds in "cada dos semanas." I will talk about B and V instead of D because D is ridiculously subtle.

In English and a bunch of other languages, B is B and V is V. But in Spanish, it all depends on where the letter appears in relation to other sounds. If either of these letters is at the beginning of an utterance, or right after L, M, N or S, it tends to be pronounced like the B in English. Otherwise they tend to be pronounced "soft," that is, like the V in English. However, there is regional and individual variation. For example:


will often be pronounced as



Cien vacas

would be pronounced by some as

Cien bacas

Here is a sample sentence:

La vaca no vale nada

Here, because "vaca" and "vale" are buried in the middle of the phrase, they will both be soft (V).

  • 2
    Note that if you said cien vacas most speakers will pronounce v as a stop, a bit like ciembacas. So it's not whether it's at the beginning of the phrase or not, but rather, what (if anything) is before it. As with French, a nasal consonant blocks lenition.
    – pablodf76
    Commented Oct 4, 2018 at 16:58
  • I'd just like to point out something I've learned from my own observations with pronunciation of the "b" and the "v" in Spanish. In an effort to improve my pronunciation, I have used tools that graph out the sound uttered by native speakers using words containing these two letters (and then compare it to my own). What I have discovered is that the advice given to beginning students of Spanish can vary. This page here, is a good example of that.
    – Lisa Beck
    Commented Sep 6, 2019 at 3:30
  • All pronunciation tips aside, what I have learned is that the Spanish "v" is a whole lot closer to the English "v" than some would have you believe. When I keep this in mind, my audio graphs more closely approximate those of the native speaker. Just my two cents. Obviously, I'm not including too much science or linguistic terminology in this comment. I merely add it to help others learn from the mistakes I now can see and hear I've been making.
    – Lisa Beck
    Commented Sep 6, 2019 at 3:30
  • @LisaBeck - Wait, who told you not to pronounce Spanish v like English v? // Consider the sentence, "Los ábacos ya no se usan tanto en China como antes." Many people will pronounce this more like "ávacos" than ábacos. And some people will pronounce "vale" like "bale." It's important that you know this, to support your understanding of spoken Spanish. It's not important that you have two versions of B and two versions of V in your own pronunciation repertoire. That is not necessary. People will understand you just fine without that. Commented Sep 7, 2019 at 6:22
  • No one told me not to pronunce the "v" like a "v," but I have seen this advice on the web (or advice that might incline one to "overcompensate" to the point of losing the qualities of the "v" sound). NTL, I concede that the sound of the "v" may depend on various aspects such as the placement of this letter w/in a sentence or word, the region, and even the speaker. As for who is dispensing this advice, various sites (and I'm not even disputing what they claim because I haven't studied this topic all that extensively), I'll link you to some in my next comment.
    – Lisa Beck
    Commented Sep 19, 2019 at 8:03

Yes, enlace is the term in Spanish that Spanish Linguists use, it is about the same system as French where consonants on the end of words will make open syllables with words that begin in a vowel, two vowels, two weaks or a strong and a weak, will 'dipthongize'(Apply your knowledge of Semi-consonants vs Semi-vowels here) and with two of the same consonant next to each-other, in which you can eliminate the first one and create a new open syllable with the consonant left.

Also when a "d" or its sound approximation is between two vowels in Spanish it can tend to disappear or soften such as in the case of a Rosalía song Bagdad where she sings Senta'íta which comes from the word sentadita, in this case the d softens between the two vowels and the i has to accented to keep the pronunciation from creating a dipthong.


I hear a near(not involving true silent consonants) liason unofficially everywhere where consonants or vowels at the beginning of words combine with the ends of the previous words. For instance here is a list I just memorized: I hear “densa yamplia”, but when I look up the transcript it is "densa y amplia". “muchos fwerzoh” is really "mucho esfuerzo", “contaborgooyosah” is contaba orgullosa, “ahtah dah la seeyah” is atada a la silla, “sufreemee entos ecoloheecos” is "sufrimientos psicológicos", “santow deeablah” is "Santa o diabla", “salsoh” is "se alzó", “sereeah pas eeguada” is "sería apaciguada," “dee see poolomado” is really "discípulo amado,"

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.